Optimum SkyNet Engine HDI

2021: Achieve 98% accuracy of pattern recognition in a mobile device

On February 19, 2021, it became known that the resident of the cluster information technologies of the fund "" Skolkovo- LLC "" Institute for the Development of the Digital Economy released an updated version of the digital platform " recognitions of images SkyNet Angne" (SkyNet Engine) using technologies. artificial intelligence The company's specialists optimized the speed and accuracy of recognizing goods on store shelves from photos directly mobile in the device. Now, in less than a second, it is possible to achieve recognition accuracy of 98%, even on ordinary and inexpensive ones. smartphones

SkyNet Engine

As explained, SkyNet Engine is a platform for machine learning models, based on its own recognition techniques and kernels. According to Hussein Az-zari, General Director of IRCE LLC, this is an image recognition system that analyzes an image in a smartphone or on a tablet without an Internet connection.

The SkyNet Engine can be used to develop image recognition systems across industries. The solution is already used for retail, trade allowing you to check the availability and quality of the placement of goods, analyze the fraction of the shelf for each category of goods, and timely update the price tags on the shelves. The system also analyzes POS materials and specialized stands in retail outlets. The SkyNet Retail image recognition software based on the SkyNet Engine platform optimizes the work of merchandisers, auditors and sales representatives.

SkyNet Engine

In industry, the recognition platform can be used to develop quality control systems on the conveyor or monitor equipment malfunctions. In healthcare - for the primary diagnosis of some diseases by photography and for the analysis of medical images. In agriculture, the system will help detect pests, notice plant diseases, control the ripening of the crop.

The field of application of our "engine" is extensive: it can be embedded in mobile applications and video surveillance systems to solve various problems in many sectors of the economy. For example, based on the SkyNet Engine, you can make an application that identifies employees to access the room. Foreign retail chains also use the solution to monitor user mood. Domestic retailers are also moving in this direction: an assessment of the emotional state of customers is carried out when tasting products. noted Hussein Al-Dawn, General Director of IRCE LLC

SkyNet Engine

The SkyNet Engine platform is included in the Unified Register of Russian Programs for Electronic Computers and Databases of the Ministry of Communications and Mass Communications of the Russian Federation (No. 6078).

In 2019, the international company Mars began the introduction of an automatic recognition system for goods using artificial intelligence on the SkyNet Engine platform in all countries of the Central Eurasia, Belarus and Turkey region.

Thanks to technological capabilities, the SkyNet Engine solution allowed Mars to obtain reliable marketing performance at any outlet in a matter of seconds, despite the quality of mobile communications. As a result of the introduction, the working hours of merchandisers in retail outlets decreased by about 25%, which ensured an increase in coverage of the territory and increased the presence of Mars in the region.

More than 70% of the data that affects point-of-sale performance and sales are automatically populated based on the analysis of photos in the mobile device. With the SkyNet Engine image recognition platform, Mars receives 5 times more marketing data than before. Now, the evaluation of the representation of goods is carried out in the context of each product position (SKU) both for its own products and for the products of local and international competitors.

As of February 2021, more than 800 mobile employees of Mars are working with the SkyNet Retail recognition system in 7 countries of the region (Kazakhstan, Belarus, Kyrgyzstan, Uzbekistan, Armenia, Georgia, Azerbaijan). In the coming year, implementation is planned in other countries of the region - in Moldova, Mongolia, Tajikistan, Turkmenistan and Turkey.

2019: Development of Optimum SkyNet Engine. Inclusion in the Register of Domestic Software

On December 17, 2019, a resident of the Skolkovo Foundation's information technology cluster, the Institute for the Development of the Digital Economy (IRCE), announced the development of a pattern recognition system on a mobile device and its registration with the Russian Ministry of Communications. The fundamental difference between development and analogues is that the full recognition cycle is available offline on smartphones and tablets.

According to the order of the Ministry of Digital Development, Communications and Mass Communications of the Russian Federation No. 742, the digital platform "OPTIMAL SkyNet Engine," as well as the image recognition system in retail "OPTIMAL SKYNet Retail" (OPTIMUM SkyNet Retail) on its basis, were entered into the Unified Register of domestic software under numbers 6078 and 6080, respectively.

According to the developer, the digital platform "OPTIMAL SKYNet Angne" recognizes and analyzes images on a mobile device using technologies artificial intelligence (mobile). neuronets In a few seconds, it identifies analyzes objects with an accuracy of more than 95%, analyzes them for compliance with the specified parameters and issues recommendations for correcting shortcomings. The platform does not require a network connection, which Internet distinguishes development from analogues. On the basis of the OPTIMAL SKYNet Angne platform, solutions work for various sectors of the economy, in particular retail and. industries

OPTIMUM SKYNet Retail software system based on OPTIMUM SKYNet Eng performs a number of functions for trading. For example, it recognizes goods and prices, analyzes the layout for corporate standards, checks the relevance of prices and current stocks, makes recommendations for correcting errors, automatically fills in key performance indicators for working with the outlet and creates reports on selected criteria, according to the HDI.

According to the developer, OPTIMAL SKYNet Retail allows you to quickly collect statistics on both your own products and the representation of competitors' goods, analyze the share of the "shelf" in the dynamics. As a result, the use of the solution reduces the working time of the merchandiser at each point, which allows you to optimize the number of personnel for servicing the retail network, as well as improves the quality and reliability of information for subsequent market analysis.

"Usually, the process of recognizing photos taken on a mobile device takes place on remote servers where images are transferred over the Internet. The "decrypted" information is then returned to the mobile device. We have developed a completely autonomous solution using a mobile neural network, which is independent of communication channels and does not require expensive servers. You can recognize and analyze photos, generate reports and recommendations using OPTIMAL SKYNet Retail immediately in the application on mobile devices, " noted' Hussein Al-zari, CEO of IRCE LLC '

According to him, industry solutions based on the OPTIMAL SKYNet Angne platform increase business efficiency, allowing enterprises to save significant amounts.